The Vision AI Label Reader developed by collective mind GmbH (COMI) is revolutionising logistics and goods-in processes by enhancing label handling capabilities. This innovative system automatically captures and reads label information, irrespective of layout, language, or code type, resulting in improved data quality, process reliability, and traceability within logistics operations.
The electronics industry faces growing challenges, as numerous components arrive from various manufacturers with increasingly complex label layouts, multilingual markings, and compressed throughput times. Previously manageable manual tasks have become bottlenecks, exacerbated by damaged barcodes or reflective packaging which increase error rates.
AI-based image processing system
COMI's Vision AI Label Reader provides a solution to this complexity by automating the capture and analysis of item information. By using an AI-based image processing system specifically designed for industrial use, it streamlines workflows and enhances data quality. The system relies on a uEye CP industrial camera from IDS Imaging Development Systems GmbH, which delivers essential image data for analysis.
This system is particularly effective for electronics manufacturing service providers
This system is particularly effective for electronics manufacturing service providers and companies with intricate logistics processes. A practical implementation is seen at Rutronik Elektronische Bauelemente GmbH, a distributor of electronic components, where it is operationally successful. It aims to autonomously gather all pertinent item details and organise them into a structured format for users.
Advanced label recognition
The Vision AI Label Reader automatically identifies significant product information and presents it systematically. It recognises all labels on an item, deciphers printed text, 1D, and 2D codes, and uses artificial intelligence for interpretation. It also processes handwritten content if needed, thus allowing it to adapt without retraining despite new layouts or languages, ensuring scalability.
A predominant feature of this system is the use of a uEye CP camera, which captures high-resolution images of labels and packaging, proving effective even under adverse conditions like reflective surfaces. This capability, supported by a well-coordinated lighting concept, results in consistently accurate recognition performance.
Compact and robust camera design
After capturing images, the AI system analyses the data through multiple stages
The camera, confined within a compact magnesium housing measuring 29 × 29 × 29 mm and weighing approximately 50 g, is equipped with the IMX183 rolling shutter CMOS sensor from Sony’s STARVIS series. Tobias Husemann, Senior Consultant at COMI, states that this sensor's back-side illumination (BSI) technology, alongside a high resolution and quick frame rate, guarantees precise detail capture even in low-light scenarios.
After capturing images, the AI system analyses the data through multiple stages, extracting and interpreting content to assign information such as part numbers and batch details. These results are directly integrated into ERP systems like SAP, enhancing real-time validation and comparison. The implementation offers significant reductions in manual checks, improves data quality, and delivers comprehensive documentation of item movements.
Boost in process efficiency
Compared to traditional multi-label readers, implementation of the Vision AI Label Reader shows a 30 per cent enhancement in efficiency. Through automation, personnel can be optimised, goods-in bottlenecks alleviated, and process integrity bolstered via early error detection.
The market is evolving towards more automated item capture solutions
The market is evolving towards more automated item capture solutions, and the Vision AI Label Reader is expected to transition from a tabletop scanner to fully integrated systems in automated warehouses. According to Husemann, systems must handle diverse lighting conditions and require a broad depth of field to remain effective across varying presentation heights and distances.
Enhancing quality control
Future developments for the 'Label Reader' involve expanded functionalities, including anomaly and defect detection, such as identifying damaged labels or defective items.
This transition will turn AI-based image processing into a core quality and inspection tool within goods-in operations, thereby playing a crucial role in maintaining order and efficiency.
The Vision AI Label Reader automatically captures and interprets label information regardless of layout, language or code type, improving process reliability, data quality and traceability in goods-in and logistics operations.
Goods-in operations in the electronics industry are under increasing pressure. Countless components from a wide range of manufacturers arrive with constantly changing label layouts, multilingual markings and ever shorter throughput times. What could once be managed manually has now become a bottleneck. Damaged barcodes or reflective packaging further increase effort and make processes more error-prone.
Image processing system
The Vision AI Label Reader from collective mind GmbH (COMI) demonstrates how this complexity can be managed. The AI-based image processing system automates the capture and interpretation of item information in goods-in and logistics – regardless of layout, language or code type. Designed for industrial use, the solution improves process reliability, enhances data quality and streamlines workflows. A uEye CP industrial camera from IDS Imaging Development Systems GmbH provides the image data required for analysis.
The Vision AI Label Reader is designed for applications where a wide variety of items, labels and packaging are processed on a daily basis. This makes it particularly suitable for electronics manufacturing service providers as well as companies with complex logistics processes and extensive inventories. One concrete example is Rutronik Elektronische Bauelemente GmbH, a globally broad-line distributor of electronic components, where the system is already in successful operation. The goal is to automatically capture all relevant item information and make it available in a structured format.
Relevant product information
The Vision AI Label Reader automatically captures all relevant product information and presents it in a structured format.
To achieve this, the system recognises all labels on an object, reads printed text as well as 1D and 2D codes, and then interprets the content using artificial intelligence. Handwritten entries can also be processed if required. Crucially, recognition does not rely on predefined label standards. New layouts, languages or code formats can be handled without retraining – a key factor for scalability and long-term viability.
Coordinated lighting concept
A central component of the solution is the industrial camera from the uEye CP family by IDS. It captures labels and packaging surfaces at high resolution and supplies the image data for AI analysis, reliably detecting fine details even under challenging conditions.
In practice, reflective packaging such as dry packs, damaged codes or fluctuating lighting conditions place high demands on image acquisition. In combination with a coordinated lighting concept, however, the system achieves consistently stable recognition performance. The use of a standard vision interface (USB3 Vision) also simplifies connection to industrial PCs and ensures easy integration into existing systems.
Compact magnesium housing
The compact magnesium housing of the uEye CP (29 × 29 × 29 mm) is both lightweight and robust, weighing around 50 g. COMI uses a model equipped with the light-sensitive IMX183 rolling shutter CMOS sensor from Sony’s STARVIS series.
Thanks to back-side illumination (BSI) technology, it delivers reliable image quality even in low-light conditions. “With a resolution of 20.44 megapixels and a frame rate of almost 20 frames per second, the camera provides exactly the level of detail we need to reliably capture even very small label information,” explains Tobias Husemann, Senior Consultant at COMI.
Stricter regulatory requirements
Following image acquisition, the AI analyses the data in several stages: Labels are localised, contents extracted and then semantically interpreted, for example to clearly assign part numbers, batches or manufacturer information. The results are transferred directly to connected ERP systems such as SAP or proALPHA, including real-time comparison and validation.
For users, this means a significant reduction in manual inspection steps and sources of error. At the same time, data quality improves and complete documentation of all item movements is created. The resulting 100 per cent traceability is increasingly becoming a decisive differentiator, particularly in view of stricter regulatory requirements in downstream industries such as medical technology.
Multi-label readers
Compared with conventional multi-label readers, practical use shows an efficiency gain of around 30 per cent in item capture. Processes can be accelerated, personnel resources deployed more effectively and bottlenecks in goods-in reduced. Automated plausibility checks of label content also increase process reliability and help identify errors at an early stage.
The market is clearly moving towards highly automated item capture. In future, the Vision AI Label Reader is set to move beyond use as a tabletop scanner and become fully integrated into automated warehouse and material flow solutions. This is already being planned in collaboration with system integrators.
Material flow solutions
According to Husemann, this also increases the demands placed on camera technology: “It has to cope with changing and sometimes unfavourable lighting conditions and work reliably on reflective surfaces. At the same time, a large depth of field is required, as labels and packaging are presented at different heights and distances and still need to be captured reliably.”
In addition, the functional scope of the ‘Label Reader’ is to be expanded gradually. Alongside pure item capture, topics such as anomaly and defect detection are coming into focus – for example identifying damaged labels, adhesive residues or defective items. This transforms AI-based image processing from a capture system into a central quality and inspection tool in goods-in. After all, tidiness is half the battle.
